In this notebook, some template code has already been provided for you, and you will need to implement additional functionality to successfully complete this project. You will not need to modify the included code beyond what is requested. Sections that begin with '(IMPLEMENTATION)' in the header indicate that the following block of code will require additional functionality which you must provide. Instructions will be provided for each section, and the specifics of the implementation are marked in the code block with a 'TODO' statement. Please be sure to read the instructions carefully!
Note: Once you have completed all of the code implementations, you need to finalize your work by exporting the Jupyter Notebook as an HTML document. Before exporting the notebook to html, all of the code cells need to have been run so that reviewers can see the final implementation and output. You can then export the notebook by using the menu above and navigating to File -> Download as -> HTML (.html). Include the finished document along with this notebook as your submission.
In addition to implementing code, there will be questions that you must answer which relate to the project and your implementation. Each section where you will answer a question is preceded by a 'Question X' header. Carefully read each question and provide thorough answers in the following text boxes that begin with 'Answer:'. Your project submission will be evaluated based on your answers to each of the questions and the implementation you provide.
Note: Code and Markdown cells can be executed using the Shift + Enter keyboard shortcut. Markdown cells can be edited by double-clicking the cell to enter edit mode.
The rubric contains optional "Stand Out Suggestions" for enhancing the project beyond the minimum requirements. If you decide to pursue the "Stand Out Suggestions", you should include the code in this Jupyter notebook.
In this notebook, you will make the first steps towards developing an algorithm that could be used as part of a mobile or web app. At the end of this project, your code will accept any user-supplied image as input. If a dog is detected in the image, it will provide an estimate of the dog's breed. If a human is detected, it will provide an estimate of the dog breed that is most resembling. The image below displays potential sample output of your finished project (... but we expect that each student's algorithm will behave differently!).

In this real-world setting, you will need to piece together a series of models to perform different tasks; for instance, the algorithm that detects humans in an image will be different from the CNN that infers dog breed. There are many points of possible failure, and no perfect algorithm exists. Your imperfect solution will nonetheless create a fun user experience!
We break the notebook into separate steps. Feel free to use the links below to navigate the notebook.
Make sure that you've downloaded the required human and dog datasets:
Note: if you are using the Udacity workspace, you DO NOT need to re-download these - they can be found in the /data folder as noted in the cell below.
Download the dog dataset. Unzip the folder and place it in this project's home directory, at the location /dog_images.
Download the human dataset. Unzip the folder and place it in the home directory, at location /lfw.
Note: If you are using a Windows machine, you are encouraged to use 7zip to extract the folder.
In the code cell below, we save the file paths for both the human (LFW) dataset and dog dataset in the numpy arrays human_files and dog_files.
!curl -O https://raw.githubusercontent.com/udacity/workspaces-student-support/master/jupyter/workspace_utils.py
import numpy as np
from glob import glob
# load filenames for human and dog images
human_files = np.array(glob("/data/lfw/*/*"))
dog_files = np.array(glob("/data/dog_images/*/*/*"))
# print number of images in each dataset
print('There are %d total human images.' % len(human_files))
print('There are %d total dog images.' % len(dog_files))
dog_files
human_files
In this section, we use OpenCV's implementation of Haar feature-based cascade classifiers to detect human faces in images.
OpenCV provides many pre-trained face detectors, stored as XML files on github. We have downloaded one of these detectors and stored it in the haarcascades directory. In the next code cell, we demonstrate how to use this detector to find human faces in a sample image.
import cv2
import matplotlib.pyplot as plt
%matplotlib inline
# extract pre-trained face detector
face_cascade = cv2.CascadeClassifier('haarcascades/haarcascade_frontalface_alt.xml')
# load color (BGR) image
img = cv2.imread(human_files[0])
# convert BGR image to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# find faces in image
faces = face_cascade.detectMultiScale(gray)
# print number of faces detected in the image
print('Number of faces detected:', len(faces))
# get bounding box for each detected face
for (x,y,w,h) in faces:
# add bounding box to color image
cv2.rectangle(img,(x,y),(x+w,y+h),(255,0,0),2)
# convert BGR image to RGB for plotting
cv_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
# display the image, along with bounding box
plt.imshow(cv_rgb)
plt.show()
Before using any of the face detectors, it is standard procedure to convert the images to grayscale. The detectMultiScale function executes the classifier stored in face_cascade and takes the grayscale image as a parameter.
In the above code, faces is a numpy array of detected faces, where each row corresponds to a detected face. Each detected face is a 1D array with four entries that specifies the bounding box of the detected face. The first two entries in the array (extracted in the above code as x and y) specify the horizontal and vertical positions of the top left corner of the bounding box. The last two entries in the array (extracted here as w and h) specify the width and height of the box.
We can use this procedure to write a function that returns True if a human face is detected in an image and False otherwise. This function, aptly named face_detector, takes a string-valued file path to an image as input and appears in the code block below.
# returns "True" if face is detected in image stored at img_path
def face_detector(img_path):
img = cv2.imread(img_path)
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray)
return len(faces) > 0
Question 1: Use the code cell below to test the performance of the face_detector function.
human_files have a detected human face? dog_files have a detected human face? Ideally, we would like 100% of human images with a detected face and 0% of dog images with a detected face. You will see that our algorithm falls short of this goal, but still gives acceptable performance. We extract the file paths for the first 100 images from each of the datasets and store them in the numpy arrays human_files_short and dog_files_short.
Answer: (You can print out your results and/or write your percentages in this cell)
98% Humans classified correctly
17% Dogs classified correctly
from tqdm import tqdm
human_files_short = human_files[:100]
dog_files_short = dog_files[:100]
#-#-# Do NOT modify the code above this line. #-#-#
## TODO: Test the performance of the face_detector algorithm
## on the images in human_files_short and dog_files_short.
human_counter = []
dog_counter = []
for i in tqdm(range(100)):
human_counter.append(face_detector(human_files_short[i]))
dog_counter.append(face_detector(dog_files_short[i]))
print('{:.0f}% Humans classified correctly'.format(sum(human_counter)))
print('{:.0f}% Dogs classified incorrectly'.format(sum(dog_counter)))
We suggest the face detector from OpenCV as a potential way to detect human images in your algorithm, but you are free to explore other approaches, especially approaches that make use of deep learning :). Please use the code cell below to design and test your own face detection algorithm. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.
### (Optional)
### TODO: Test performance of anotherface detection algorithm.
### Feel free to use as many code cells as needed.
In this section, we use a pre-trained model to detect dogs in images.
The code cell below downloads the VGG-16 model, along with weights that have been trained on ImageNet, a very large, very popular dataset used for image classification and other vision tasks. ImageNet contains over 10 million URLs, each linking to an image containing an object from one of 1000 categories.
import torch
import torchvision.models as models
# define VGG16 model
VGG16 = models.vgg16(pretrained=True)
# check if CUDA is available
use_cuda = torch.cuda.is_available()
# move model to GPU if CUDA is available
if use_cuda:
VGG16 = VGG16.cuda()
Given an image, this pre-trained VGG-16 model returns a prediction (derived from the 1000 possible categories in ImageNet) for the object that is contained in the image.
In the next code cell, you will write a function that accepts a path to an image (such as 'dogImages/train/001.Affenpinscher/Affenpinscher_00001.jpg') as input and returns the index corresponding to the ImageNet class that is predicted by the pre-trained VGG-16 model. The output should always be an integer between 0 and 999, inclusive.
Before writing the function, make sure that you take the time to learn how to appropriately pre-process tensors for pre-trained models in the PyTorch documentation.
from PIL import Image
import torchvision.transforms as transforms
def VGG16_predict(img_path):
'''
Use pre-trained VGG-16 model to obtain index corresponding to
predicted ImageNet class for image at specified path
Args:
img_path: path to an image
Returns:
Index corresponding to VGG-16 model's prediction
'''
## TODO: Complete the function.
## Load and pre-process an image from the given img_path
## Return the *index* of the predicted class for that image
# open image
img = Image.open(img_path)
# define transform required for vgg16
transform = transforms.Compose([transforms.Resize(256),
transforms.CenterCrop(224),
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406),
(0.229, 0.224, 0.225))])
# transform image
img_t = transform(img)
# add batch dimension
batch_t = img_t.unsqueeze(0)
if use_cuda:
batch_t = batch_t.cuda()
# set model to evaluation
VGG16.eval()
# run inference
out = VGG16(batch_t)
# get prediction through max value
prediction = out.cpu().detach().numpy().argmax()
return prediction # predicted class index
VGG16_predict(dog_files[156])
While looking at the dictionary, you will notice that the categories corresponding to dogs appear in an uninterrupted sequence and correspond to dictionary keys 151-268, inclusive, to include all categories from 'Chihuahua' to 'Mexican hairless'. Thus, in order to check to see if an image is predicted to contain a dog by the pre-trained VGG-16 model, we need only check if the pre-trained model predicts an index between 151 and 268 (inclusive).
Use these ideas to complete the dog_detector function below, which returns True if a dog is detected in an image (and False if not).
### returns "True" if a dog is detected in the image stored at img_path
def dog_detector(img_path):
## TODO: Complete the function.
prediction = VGG16_predict(img_path)
return (151 <= prediction <= 268)# true/false
dog_detector(human_files[199])
dog_detector(dog_files[199])
Question 2: Use the code cell below to test the performance of your dog_detector function.
human_files_short have a detected dog? dog_files_short have a detected dog?Answer:
100% Dogs classified correctly
100% Humans classified correctly
### TODO: Test the performance of the dog_detector function
### on the images in human_files_short and dog_files_short.
from tqdm import tqdm
dog_is_dog = []
human_is_dog = []
for i in tqdm(range(100)):
dog_is_dog.append(dog_detector(dog_files_short[i]))
human_is_dog.append(dog_detector(human_files_short[i]))
print('{:.0f}% Dogs classified correctly'.format(sum(dog_is_dog)))
print('{:.0f}% Humans classified incorrectly'.format(sum(human_is_dog)))
We suggest VGG-16 as a potential network to detect dog images in your algorithm, but you are free to explore other pre-trained networks (such as Inception-v3, ResNet-50, etc). Please use the code cell below to test other pre-trained PyTorch models. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.
### (Optional)
### TODO: Report the performance of another pre-trained network.
### Feel free to use as many code cells as needed.
Now that we have functions for detecting humans and dogs in images, we need a way to predict breed from images. In this step, you will create a CNN that classifies dog breeds. You must create your CNN from scratch (so, you can't use transfer learning yet!), and you must attain a test accuracy of at least 10%. In Step 4 of this notebook, you will have the opportunity to use transfer learning to create a CNN that attains greatly improved accuracy.
We mention that the task of assigning breed to dogs from images is considered exceptionally challenging. To see why, consider that even a human would have trouble distinguishing between a Brittany and a Welsh Springer Spaniel.
| Brittany | Welsh Springer Spaniel |
|---|---|
![]() |
![]() |
It is not difficult to find other dog breed pairs with minimal inter-class variation (for instance, Curly-Coated Retrievers and American Water Spaniels).
| Curly-Coated Retriever | American Water Spaniel |
|---|---|
![]() |
![]() |
Likewise, recall that labradors come in yellow, chocolate, and black. Your vision-based algorithm will have to conquer this high intra-class variation to determine how to classify all of these different shades as the same breed.
| Yellow Labrador | Chocolate Labrador | Black Labrador |
|---|---|---|
![]() |
![]() |
![]() |
We also mention that random chance presents an exceptionally low bar: setting aside the fact that the classes are slightly imabalanced, a random guess will provide a correct answer roughly 1 in 133 times, which corresponds to an accuracy of less than 1%.
Remember that the practice is far ahead of the theory in deep learning. Experiment with many different architectures, and trust your intuition. And, of course, have fun!
Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dog_images/train, dog_images/valid, and dog_images/test, respectively). You may find this documentation on custom datasets to be a useful resource. If you are interested in augmenting your training and/or validation data, check out the wide variety of transforms!
import os
from torchvision import datasets
### TODO: Write data loaders for training, validation, and test sets
## Specify appropriate transforms, and batch_sizes
batch_size = 20
# define a transormation with data augmentation
transform = transforms.Compose([transforms.Resize(256),
transforms.CenterCrop(224),
transforms.RandomHorizontalFlip(),
transforms.RandomRotation(10),
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406),
(0.229, 0.224, 0.225))])
transform_v_t = transforms.Compose([
transforms.Resize(256),
transforms.CenterCrop(224),
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406),
(0.229, 0.224, 0.225))
])
# '/data/dog_images/train/103.Mastiff/Mastiff_06833.jpg'
# load data
train_data = datasets.ImageFolder('/data/dog_images/train', transform=transform)
valid_data = datasets.ImageFolder('/data/dog_images/valid', transform=transform_v_t)
test_data = datasets.ImageFolder('/data/dog_images/test', transform=transform_v_t)
# prepare dataloaders
train_loader = torch.utils.data.DataLoader(train_data, batch_size=batch_size, shuffle=True)
valid_loader = torch.utils.data.DataLoader(valid_data, batch_size=batch_size, shuffle=True)
test_loader = torch.utils.data.DataLoader(test_data, batch_size=batch_size, shuffle=True)
import matplotlib.pyplot as plt
import numpy as np
%matplotlib inline
# helper function to unormalize an image and display it
def imshow(img):
# define inverse normalization
inv_normalize = transforms.Normalize(
mean=[-0.485/0.229, -0.456/0.224, -0.406/0.225],
std=[1/0.229, 1/0.224, 1/0.225])
# apply inverse normalization
img = inv_normalize(img)
# convert to numpy array for display
img = img.numpy()
# convert from Tensor image and show
plt.imshow(np.transpose(img, (1,2,0)))
# obtain one batch of training images
dataiter = iter(train_loader)
images, labels = dataiter.next()
# plot the images in the batch, along with the corresponding labels
fig = plt.figure(figsize=(25,4))
# display 20 images
for idx in np.arange(20):
ax = fig.add_subplot(2, 20/2, idx+1, xticks=[], yticks=[])
imshow(images[idx])
ax.set_title(labels[idx].item())
labels
Question 3: Describe your chosen procedure for preprocessing the data.
Answer:
Create a CNN to classify dog breed. Use the template in the code cell below.
import torch.nn as nn
import torch.nn.functional as F
# define the CNN architecture
class Net(nn.Module):
### TODO: choose an architecture, and complete the class
def __init__(self):
super(Net, self).__init__()
## Define layers of a CNN
# conv layer sees 224x224x3 image tensor
self.conv1 = nn.Conv2d(3, 32, 3, padding=1)
self.bn1 = nn.BatchNorm2d(32)
# conv layer sees 112x112x32 image tensor
self.conv2 = nn.Conv2d(32, 128, 3, padding=1)
self.bn2 = nn.BatchNorm2d(128)
# conv layer sees 56x56x128 image tensor
self.conv3 = nn.Conv2d(128, 256, 3, padding=1)
self.bn3 = nn.BatchNorm2d(256)
# conv layer sees 28x28x256 image tensor
self.conv4 = nn.Conv2d(256, 740, 3, padding=1)
self.bn4 = nn.BatchNorm2d(740)
# conv layer sees 28x28x740 image tensor
# max pooling layer
self.pool = nn.MaxPool2d(2,2)
# linear layer sees 14x14x740 -> 1000
self.fc1 = nn.Linear(14*14*740, 1000)
# linear layer (1000 -> 500)
self.fc2 = nn.Linear(1000, 500)
# linear layer (500 -> 133) output dog classes
self.fc3 = nn.Linear(500, 133)
# droput layer p=0.25
self.drop = nn.Dropout(0.45)
def forward(self, x):
## Define forward behavior
# sequence of convolutional and pooling layers
x = self.pool(F.relu(self.bn1(self.conv1(x))))
x = self.pool(F.relu(self.bn2(self.conv2(x))))
x = self.pool(F.relu(self.bn3(self.conv3(x))))
x = self.pool(F.relu(self.bn4(self.conv4(x))))
#x = F.relu(self.conv5(x))
# flatten image input
x = x.view(-1, 14*14*740)
# add dropout layer
x = self.drop(x)
# add first hidden layer, with relu activation layer
x = F.relu(self.fc1(x))
# add dropout layer
x = self.drop(x)
# add second hidden layer, with relu activation layer
x = F.relu(self.fc2(x))
# add dropout layer
x = self.drop(x)
# add final hidden layer with output classes
x = self.fc3(x)
return x
#-#-# You do NOT have to modify the code below this line. #-#-#
# instantiate the CNN
model_scratch = Net()
# move tensors to GPU if CUDA is available
if use_cuda:
model_scratch.cuda()
print(model_scratch)
Question 4: Outline the steps you took to get to your final CNN architecture and your reasoning at each step.
Answer:
I followed a similar structure to what we learned in the cifar10 lesson, but extended the convolutional layers to follow the depth used by vgg16. I kept it smaller than vgg16 to speed up training.
Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_scratch, and the optimizer as optimizer_scratch below.
import torch.optim as optim
### TODO: select loss function
criterion_scratch = nn.CrossEntropyLoss()
### TODO: select optimizer
optimizer_scratch = optim.Adam(model_scratch.parameters())
Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_scratch.pt'.
# got an error after training: OSError: image file is truncated (150 bytes not processed)
# online solution said to paste the code below
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True
# define loaders dictionary storing the dataloaders
loaders_scratch = {'train': train_loader, 'valid': valid_loader, 'test': test_loader}
def train(n_epochs, loaders, model, optimizer, criterion, use_cuda, save_path):
"""returns trained model"""
# initialize tracker for minimum validation loss
valid_loss_min = np.Inf
for epoch in range(1, n_epochs+1):
# initialize variables to monitor training and validation loss
train_loss = 0.0
valid_loss = 0.0
###################
# train the model #
###################
model.train()
for batch_idx, (data, target) in enumerate(loaders['train']):
# move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
## find the loss and update the model parameters accordingly
# set grads to zero
optimizer.zero_grad()
# forward computation
out = model(data)
# calculate loss
loss = criterion(out, target)
# backward pass: compute gradient of the loss with respect to model parameters
loss.backward()
# perform a single optimization step (parameter update)
optimizer.step()
## record the average training loss, using something like
train_loss = train_loss + ((1 / (batch_idx + 1)) * (loss.data - train_loss))
######################
# validate the model #
######################
model.eval()
for batch_idx, (data, target) in enumerate(loaders['valid']):
# move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
# forward pass: compute predicted outputs by passing inputs to the model
out = model(data)
# calculate the batch loss
loss = criterion(out, target)
## update the average validation loss
valid_loss = valid_loss + ((1 / (batch_idx + 1)) * (loss.data - valid_loss))
# print training/validation statistics
print('Epoch: {} \tTraining Loss: {:.6f} \tValidation Loss: {:.6f}'.format(
epoch,
train_loss,
valid_loss
))
## TODO: save the model if validation loss has decreased
if valid_loss <= valid_loss_min:
print('Validation loss decreased ({:.6f} -> {:.6f}). Saving model...'.format(valid_loss_min, valid_loss))
# save model
torch.save(model.state_dict(), save_path)
valid_loss_min = valid_loss
# return trained model
return model
# then wrap long-running work in below function
from workspace_utils import active_session
# this is to keep the workspace active
with active_session():
# train the model
model_scratch = train(100, loaders_scratch, model_scratch, optimizer_scratch,
criterion_scratch, use_cuda, 'model_scratch.pt')
# load the model that got the best validation accuracy
model_scratch.load_state_dict(torch.load('model_scratch.pt'))
# load model and optimizer states to resume training
# Additional information
EPOCH = 29
PATH = "model_scratch.pt"
TLOSS = 3.734124
VLOSS = 3.752248
torch.save({
'epoch': EPOCH,
'model_state_dict': net.state_dict(),
'optimizer_state_dict': optimizer.state_dict(),
't_loss': TLOSS,
'v_loss': VLOSS,
}, PATH)
Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 10%.
def test(loaders, model, criterion, use_cuda):
# monitor test loss and accuracy
test_loss = 0.
correct = 0.
total = 0.
model.eval()
for batch_idx, (data, target) in enumerate(loaders['test']):
# move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
# forward pass: compute predicted outputs by passing inputs to the model
output = model(data)
# calculate the loss
loss = criterion(output, target)
# update average test loss
test_loss = test_loss + ((1 / (batch_idx + 1)) * (loss.data - test_loss))
# convert output probabilities to predicted class
pred = output.data.max(1, keepdim=True)[1]
# compare predictions to true label
correct += np.sum(np.squeeze(pred.eq(target.data.view_as(pred))).cpu().numpy())
total += data.size(0)
print('Test Loss: {:.6f}\n'.format(test_loss))
print('\nTest Accuracy: %2d%% (%2d/%2d)' % (
100. * correct / total, correct, total))
# call test function
# the active_session is to keep udacitys workspace active and not shut down during long term calculations
with active_session():
test(loaders_scratch, model_scratch, criterion_scratch, use_cuda)
You will now use transfer learning to create a CNN that can identify dog breed from images. Your CNN must attain at least 60% accuracy on the test set.
Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dogImages/train, dogImages/valid, and dogImages/test, respectively).
If you like, you are welcome to use the same data loaders from the previous step, when you created a CNN from scratch.
## TODO: Specify data loaders
import os
from torchvision import datasets
from PIL import Image
import torchvision.transforms as transforms
import torch
batch_size = 20
transform = transforms.Compose([
transforms.Resize(256),
transforms.CenterCrop(224),
transforms.RandomHorizontalFlip(),
transforms.RandomRotation(10),
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))])
transform_v_t = transforms.Compose([
transforms.Resize(256),
transforms.CenterCrop(224),
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406),
(0.229, 0.224, 0.225))
])
# import data
train_data = datasets.ImageFolder('/data/dog_images/train', transform=transform)
valid_data = datasets.ImageFolder('/data/dog_images/valid', transform=transform_v_t)
test_data = datasets.ImageFolder('/data/dog_images/test', transform=transform_v_t)
# prepare dataloaders
train_loader = torch.utils.data.DataLoader(train_data, batch_size=batch_size, shuffle=True)
valid_loader = torch.utils.data.DataLoader(valid_data, batch_size=batch_size, shuffle=True)
test_loader = torch.utils.data.DataLoader(test_data, batch_size=batch_size, shuffle=True)
Use transfer learning to create a CNN to classify dog breed. Use the code cell below, and save your initialized model as the variable model_transfer.
To define a model for training we'll follow these steps:
Freezing simply means that the parameters in the pre-trained model will not change during training.
import torchvision.models as models
import torch.nn as nn
## TODO: Specify model architecture
# load the pretrained model
model_transfer = models.vgg16(pretrained=True)
# freeze training for all 'features' layers
for param in model_transfer.features.parameters():
param.requires_grad = False
# check the number of parameters in the last layer
print(model_transfer.classifier[6].in_features)
print(model_transfer.classifier[6].out_features)
# replace the last fully connected layer to match the number of outputs for this problem
model_transfer.classifier[6] = nn.Linear(in_features=4096, out_features=133, bias=True)
# check if CUDA is available
use_cuda = torch.cuda.is_available()
# if cuda available, move the model to gpu
if use_cuda:
model_transfer = model_transfer.cuda()
Question 5: Outline the steps you took to get to your final CNN architecture and your reasoning at each step. Describe why you think the architecture is suitable for the current problem.
Answer:
1) Load a pretrained model
2) Freeze training parameters on features layers
3) Replace the last fully connected layer to represent the number of classes in our application (133)
I think this architecture is suitable for this project as is similar to the lesson we had on transfer learning for flower types.
Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_transfer, and the optimizer as optimizer_transfer below.
import torch.optim as optim
# categorical cross entropy
criterion_transfer = nn.CrossEntropyLoss()
# optimization (stochastic gradient descent)
optimizer_transfer = optim.Adam(model_transfer.classifier.parameters())
Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_transfer.pt'.
# data dictionary
data_transfer = {'train': train_data, 'valid': valid_data, 'test': test_data}
# loader dictionary
loaders_transfer = {'train': train_loader, 'valid': valid_loader, 'test': test_loader}
# got an error after training: OSError: image file is truncated (150 bytes not processed)
# online solution said to paste the code below
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True
# training routine
def train(n_epochs, loaders, model, optimizer, criterion, use_cuda, save_path):
# set validation loss to max
valid_loss_min = np.Inf
for epoch in range(n_epochs):
train_loss = 0.0
valid_loss = 0.0
##################
# Train the model
#################
for batch_idx, (data, target) in enumerate(loaders['train']):
# move tensors to gpu if cuda available
if use_cuda:
data, target = data.cuda(), target.cuda()
# reset gradients
optimizer.zero_grad()
# forward pass
output = model(data)
# calculate loss
loss = criterion(output, target)
# backward pass
loss.backward()
# optimization step
optimizer.step()
## record the average training loss, using something like
train_loss = train_loss + ((1 / (batch_idx + 1)) * (loss.data - train_loss))
####################
# Validate the model
####################
for batch_idx, (data, target) in enumerate(loaders['valid']):
# move tensors to gpu if cuda available
if use_cuda:
data, target = data.cuda(), target.cuda()
# compute forward pass
output = model(data)
# compute loss
loss = criterion(output, target)
# record the average valid loss
valid_loss = valid_loss + ((1/(batch_idx + 1)) * (loss.data - valid_loss))
##############################
# Save model if loss is lowest
##############################
if valid_loss < valid_loss_min:
print('Validation loss decreased ({:.6f} -> {:.6f}). Saving model...'.format(valid_loss_min, valid_loss))
# save model
torch.save(model.state_dict(), save_path)
valid_loss_min = valid_loss
# print training/validation statistics
print('Epoch: {} \tTraining Loss: {:.6f} \tValidation Loss: {:.6f}'.format(
epoch,
train_loss,
valid_loss
))
return model
# train the model
# the active_session is to keep udacitys workspace active and not shut down during long term calculations
with active_session():
model_transfer = train(40, loaders_transfer, model_transfer, optimizer_transfer, criterion_transfer, use_cuda, 'model_transfer.pt')
# load the model that got the best validation accuracy (uncomment the line below)
model_transfer.load_state_dict(torch.load('model_transfer.pt'))
Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 60%.
test(loaders_transfer, model_transfer, criterion_transfer, use_cuda)
Write a function that takes an image path as input and returns the dog breed (Affenpinscher, Afghan hound, etc) that is predicted by your model.
# extracting class names from data vars
data_transfer['train'].classes[1][4:].replace("_", " ")
### TODO: Write a function that takes a path to an image as input
### and returns the dog breed that is predicted by the model.
# list of class names by index, i.e. a name can be accessed like class_names[0]
class_names = [item[4:].replace("_", " ") for item in data_transfer['train'].classes]
def predict_breed_transfer(img_path):
# load the image and return the predicted breed
# load image
img = Image.open(img_path)
# define transform
transform = transforms.Compose([
transforms.Resize(256),
transforms.CenterCrop(224),
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406),
(0.229, 0.224, 0.225))
])
# transform image
img = transform(img)
# add batch dimension
img = img.unsqueeze(0)
# move to gpu if cuda available
if use_cuda:
img = img.cuda()
# run inference
output = model_transfer(img)
# get prediction through max value
prediction = output.cpu().detach().numpy().argmax()
return class_names[prediction]
predict_breed_transfer(dog_files[-1])
class_names
import matplotlib.pyplot as plt
import numpy as np
%matplotlib inline
# helper function to unormalize an image and display it
def imshow(img):
# define inverse normalization
inv_normalize = transforms.Normalize(
mean=[-0.485/0.229, -0.456/0.224, -0.406/0.225],
std=[1/0.229, 1/0.224, 1/0.225])
# apply inverse normalization
img = inv_normalize(img)
# convert to numpy array for display
img = img.numpy()
# convert from Tensor image and show
plt.imshow(np.transpose(img, (1,2,0)))
# obtain one batch of test images
dataiter = iter(loaders_transfer['test'])
images, labels = dataiter.next()
images.numpy()
# move model inputs to cuda, if GPU available
if use_cuda:
images = images.cuda()
# get sample outputs
output = model_transfer(images)
# convert output probabilities to predicted class
_, preds_tensor = torch.max(output, 1)
preds = np.squeeze(preds_tensor.numpy()) if not use_cuda else np.squeeze(preds_tensor.cpu().numpy())
# move model inputs back to cpu for display, if GPU available
if use_cuda:
images = images.cpu()
# plot the images in the batch, along with predicted and true labels
fig = plt.figure(figsize=(30, 6))
for idx in np.arange(20):
ax = fig.add_subplot(2, 20/2, idx+1, xticks=[], yticks=[])
imshow(images[idx])
ax.set_title("{} \n ({})".format(class_names[preds[idx]], class_names[labels[idx]]),
color=("green" if preds[idx]==labels[idx].item() else "red"))
Write an algorithm that accepts a file path to an image and first determines whether the image contains a human, dog, or neither. Then,
You are welcome to write your own functions for detecting humans and dogs in images, but feel free to use the face_detector and human_detector functions developed above. You are required to use your CNN from Step 4 to predict dog breed.
Some sample output for our algorithm is provided below, but feel free to design your own user experience!

# download class names for pretrained VGG16
import urllib, json
url = 'https://s3.amazonaws.com/deep-learning-models/image-models/imagenet_class_index.json'
response = urllib.request.urlopen(url)
# dictionary with class id, code and name
class_idx = json.loads(response.read())
# create list with class names indexed by number
idx2label = [class_idx[str(k)][1] for k in range(len(class_idx))]
# function to show image with breed
def show_result(img_path, salutation, breed):
# open image and show it
img = Image.open(img_path)
print(f'hello, {salutation}!')
plt.imshow(img)
plt.show()
print(f'You look like a {breed}...')
### TODO: Write your algorithm.
### Feel free to use as many code cells as needed.
def run_app(img_path):
## handle cases for a human face, dog, and neither
# boolean: dog detector
is_dog = dog_detector(img_path)
# boolean: human detector
is_human = face_detector(img_path)
# evaluated XOR lambda: if both are True or False then is_neither is True, else False
is_neither = (lambda x, y: True if (x + y) % 2 == 0 else False)(is_dog, is_human)
if is_neither:
show_result(img_path, 'alien', 'thing from another world')
else:
# run inference to predict breed using model trained with transfer learning
breed = predict_breed_transfer(img_path)
# run indference with pretrained vgg16 and return class name
#breed = idx2label[VGG16_predict(img_path)]
# returns the salutation, either dog or human
salutation = (lambda x: 'human' if x else 'dog')(is_human)
show_result(img_path, salutation, breed)
# debug
'''
print('Dog detected: ', is_dog)
print('Human detected: ', is_human)
print('Is neither ', is_neither)
'''
run_app(human_files[11])
run_app(dog_files[100])
print('hello, ', 'human', '!')
In this section, you will take your new algorithm for a spin! What kind of dog does the algorithm think that you look like? If you have a dog, does it predict your dog's breed accurately? If you have a cat, does it mistakenly think that your cat is a dog?
Test your algorithm at least six images on your computer. Feel free to use any images you like. Use at least two human and two dog images.
Question 6: Is the output better than you expected :) ? Or worse :( ? Provide at least three possible points of improvement for your algorithm.
Answer:
I am actually happy with the implementation and the output.. it feels efficient and correct.
One thing that could always be improved is the underlying detection algorithms, but that requires more training and loads of time.
From the results I can also see that the face detector is not very accurate as the golden retriever was cataloged as a human, however, the dog classifier did spot it correctly.
Another improvement to the algorithm could be to use the full vgg16 to classify those photos which are neither human nor of a dog.
I could also improve the way results are displayed to make it more user friendly.
## TODO: Execute your algorithm from Step 6 on
## at least 6 images on your computer.
## Feel free to use as many code cells as needed.
## suggested code, below
for file in np.hstack((human_files[:3], dog_files[:3])):
run_app(file)
# load filenames for human and dog images
new_images = np.array(glob("/home/workspace/dog_project/test_images/*"))
# print number of images in each dataset
print('There are %d total new images.' % len(new_images))
new_images
for file in new_images:
run_app(file)